Here's an explanation of why altitude gain can vary significantly. The numbers in my example may be exaggerated but that's to make a point.
Imagine you're walking on a flat surface at 1,000' elevation. A GPS isn't perfectly accurate. Every reading could be off by up to 3' in either direction.
Here's a list of successive readings: 1,000, 998, 1,001, 997, 999, 1,003,...
To the software in the GPS this looks like 0, -2, +3, -4, +2, +4. So the GPS reports a gain of 3+2+4=9 feet gain. The error in cumulative altitude gain will increase the longer you walk. Yet you've been walking on a flat surface the entire time and the gain should be zero.
Your GPS makes several readings per minute. You go for a 3 hour walk. The altitude gain readings could be way off by the time you finish.
It would be interesting if you could repeat your experiment. The same three hikes using the same GPS devices and apps. I bet they'll be significantly different from the first set.
one or more gifts or other benefits from the co-op.